As part of the Transformative Change for Sustainable Development Seminar Series, NRI was pleased to welcome Professor Andy Sumner of King’s College London to present his research, which critically examines global poverty and inequality assessment.

So much of our narrative on the trends of global and national poverty and inequality rests on large global databases drawn from household survey data. We are informed that poverty and inequality have both dropped dramatically in the Sustainable Development Era (post 1980) and that the Covid pandemic had relatively little effect on the rates of global poverty and inequality. But can these narratives stand true if challenged? Have international agencies created a political economy of over-optimism where reported progress masks the need for further action? Professor Andy Sumner and Dr Eduardo Ortiz-Juarez of King’s College, London, set out to explore the validity of these claims from the very data used to construct these global narratives. Prof Sumner presented the findings to staff and students at the Natural Resources Institute (NRI) at the University of Greenwich on Friday 19 May.

Unafraid of the challenges of big datasets, Sumner and Ortiz-Juarez have constructed a large database from international data sources. The data cover the period 1981 – 2019 and 96 percent of the global population. Sumner notes that this data does not cover the post-pandemic era, so statements about post-pandemic recovery cannot be made. Secondly, credible household survey data for India, which recently surpassed China as the most populous country, is not available after 2012 and had to be estimated (as is done in all international datasets) from older data.

Sumner’s presentation covered definitions of poverty and inequality – with the World Bank initially setting the poverty line at a dollar-a-day some years ago. Over time, this has been adjusted to US$1.90 per person per day (and updated again in the most recent iteration to US$2.15).

Sumner questions if this is a realistic level of income to live on. He and his colleagues then proceed to demonstrate the sensitivity of global estimates of poverty reduction to the inclusion of China, India, Indonesia and Vietnam. Where you draw the poverty line changes the relative contribution of different countries to global poverty counts and dramatically affects overall poverty estimates.

They show that every 10 cents added to the poverty line, moves roughly 70 million more people into poverty, demonstrating how sensitive estimates are to the level at which the poverty line is set. This also means that a small change in income can have very dramatic impacts on the number of poor people. Against this backdrop, it is not unreasonable to conclude that at least half of the world lives in absolute poverty and about 1 in 10 in extreme poverty.

The real question, then, is whether we can really say people have moved out of poverty and whether their lives have been truly transformed. Sumner and colleagues propose that $13 per capita is a more meaningful line in the sense that at this level, the probability is high that a person who has escaped poverty will stay non-poor. However, even this estimate does not consider the variability across countries in terms of what welfare measures are offered to citizens in terms of social security, healthcare benefits etc.

‘It is fair to say that as you prod around, the narrative that poverty is falling dramatically is fragile’. There has been economic growth across the world between 1981 and 2019. Prior to the pandemic, incomes increased and there was a dramatic reduction in the number of countries that are really poor. Thirty countries are still classified as very poor, taking the World Bank’s classification and not likely to emerge from poverty soon. For most of these countries, international aid largely determines the government’s functioning and the delivery of social services. However, there is a group of about 80 developing countries where economic growth has generated more domestic resources, which have largely replaced aid.

Since the mid-1990s, agriculture has been less important for jobs. Yet industrialisation has stalled and, in some countries, deindustrialisation is now evident. Jobs in the service industry have become more important for growth. However, these jobs are largely in informal services. Sumner surmises that the role of manufacturing as a driver of economic development may well be over. He questions what role technologies such as AI will play in the future and whether AI will generate new jobs or replace them.  

Likewise, the global narrative that inequality has been reducing over time for the last decade may also be a ‘sunshine’ narrative. The trends are largely influenced by China, India and other large emerging economies. Sumner posits that regardless of whether the post-pandemic recovery is weak or strong, we are likely to see a rise in global inequality.

‘Even if things are getting better, how much better are they?’ Sumner estimates that at current rates, pushing everyone over the $1.9 poverty line could take another generation. If we aim at getting everyone above $13, it could take 400 – 500 years to achieve. The current global narratives mask the need to consider the redistribution of resources through policies that transfer income, funded perhaps by a reduction in some regressive fossil fuel subsidies and this could, in principle, lead to significant reductions in both poverty and inequality.

Prof Vegard Iversen agrees: ‘What Sumner and colleagues show is that escapes from poverty from the 1980s onwards are much more marginal than acknowledged. As the estimates show, many people are just above the poverty line and at a high risk of falling back into poverty. Instead of getting sustainable and transformative poverty escapes, we see marginal and temporary escapes. Recalibration of poverty lines to a level where people’s lives are sufficiently improved also raises the more fundamental question of how this affects our thinking about the policies and interventions that are needed to alleviate poverty. What types of programmes and policies will be needed to generate robust and sustainable poverty escapes?’